This study involved patients with PDAC undergoing radical operations in a retrospective study cohort between January 2000 and June 2012 at Sun Yat-sen Memorial Hospital of Sun Yat-sen University and a prospective cohort from June 2012 to May 2013 (registration number: ChiCTR-TRC-12002548). The study was approved by the Ethics Committee of Sun Yat-sen Memorial Hospital, and all study participants provided informed written consent prior to study enrollment. Data were collected on age, sex, histological and pathological findings, lymph node metastasis, distant metastases, tumor stage (UICC 6^th edition), and clinical follow-up. Patients who died of postoperative complications and patients lost during follow-up were excluded. Overall survival (OS) was calculated from the date of operation to the date of death or last follow-up. Through searching the electronic medical record system including the clinical and pathological records of inpatients, 96 surgically treated patients diagnosed with histologically confirmed stage I or II PDAC were included.

Paraffin-embedded samples of the primary carcinomas from 96 patients were stained for RASSF6. The pathological sections were deparaffinized in xylene and rehydrated in a graded ethanol series followed by heat-induced epitope retrieval in citrate buffer (pH 6.0). Antigen retrieval was carried out in 10 mmol/L citrate buffer (pH 6.0) in a microwave oven for 15 min. The activity of endogenous peroxidase was exhausted with 3% hydrogen peroxide for 10 min at room temperature. A rabbit RASSF6 polyclonal antibody (Proteintech Group, Chicago, IL) was applied overnight at 4 °C at an optimal working concentration of 1:200. After sufficient phosphate buffered saline rinses, the sections were immunostained with goat anti-rabbit polymers. The staining was scored by two independent investigators, without knowledge of the patient outcomes, clinical features or pathological characteristics, according to the staining intensity and extent as described previously[18]. The sum of staining intensity and extent was designated as follows: 0-2, negative expression; 3-4, weak expression; and 5-6, strong expression. If a disagreement occurred, the slides were re-examined to obtain a final consensus.

The total RNA of 96 frozen primary pancreatic cancer tissues was prepared with Trizol (TRIzol, Invitrogen, United States) according to the manufacturer’s instructions and reverse transcribed into cDNA. The mRNA levels of RASSF6 and GAPDH were analyzed using the SYBR Green Realtime PCR Master Mix with gene-specific primers on the ABI 7500 Fast Real-Time PCR System, according to the manufacturer’s instructions. The primers for qRT-PCR were: RASSF6, sense 5’-AGCTGCCAGTTCTTGGAATG-3’ and antisense 5’-AGGCCAGACAGCTCTGATGT-3’; GAPDH, sense 5’-GTCCACCACCCTGTTGCTGTA-3’ and antisense 5’-CTTCAACAGCGACACCCACTC-3’. The cycle number at which the reaction crossed an arbitrarily placed threshold (Ct) was determined for each gene. The RASSF6 mRNA level for each tumor was calculated using the △Ct method: △Ct = Ct_RASSF6 - Ct_GAPDH.

The total protein content was extracted from frozen PDAC and para-tumor tissues stored in liquid nitrogen using a Whole Cell Extraction Kit (Pierce, Rockford, IL). For equal protein loading, a bicinchoninic acid protein assay kit (Pierce) was used to calculate the protein concentration in each sample. Equivalent amounts of proteins were separated by SDS-PAGE and transferred to polyvinylidene fluoride (PVDF) membranes for immunoblotting. The membranes were blocked in 5% fat-free milk for 2 h at room temperature and washed 3 times, and then the membranes were incubated with the following primary antibodies: a rabbit anti-human RASSF6 polyclonal antibody (1:1000, Proteintech Group, Chicago, IL) and a polyclonal rabbit anti-human β-actin polyclonal antibody (1:1000, Abcam, Cambridge, MA). β-actin was used as a loading control. Horseradish peroxidase-conjugated secondary antibodies (Cell Signaling Technology) and an ECL chemiluminescence kit (Pierce) were used to detect the bound antibodies.

All of the statistical analyses were performed with the SPSS 16.0 statistical software package (SPSS Inc., Chicago, IL). Frequency distributions were compared by χ² test. Continuous variables were compared using the Student’s t-test. The principal outcome measure was the length of survival as measured from the time of the original surgery. The patients alive at the time of the follow-up point were censored. The last follow-up period for patients still alive was May 2013. A Kaplan- Meier survival analysis was used to assess the significance of RASSF6 protein alone or together with other clinical factors. The significant differences between the survival curves were determined using the log-rank test. The 1-, 3- and 5-year survival rates were estimated using life tables. Variables that were significant after univariate analysis at P < 0.1 were included in multivariate analysis. Cox proportional hazards models were generated for multivariate analysis. A P value < 0.05 was considered statistically significant in a two-tailed test. The statistical methods of this study were reviewed by Professor Jing Gu from the School of Public Health of Sun Yat-sen University.

Demographic data and tumor characteristics are summarized in Table 1. The cohort of 96 patients consisted of 45 women and 51 men. The mean age at operation was 60.7 years, with a median age of 61 years and a range of 31-77. At the time of the last clinical follow-up (May 2013), 23 (24.0%) patients were alive. None of the patients died during the perioperative period because of post-operative complications. The median OS was 15.5 mo, with 1-, 3-, and 5-year survival rates of 57.5%, 23.0%, and 14.0%, respectively, in all 96 patients. The majority of tumors were moderately differentiated (38.5%), followed by well differentiated (36.5%), and 25.0% of tumors were poorly differentiated. Most of the tumors (n = 81) were located in the head of the pancreas (84.4%), and 80 (83.3%) were more than 2.0 cm in maximal diameter. Eighty-one (84.4%) patients had T3 tumors. Lymph node metastases were present in 61 (63.5%) patients, and portal vein invasion was present in 12 (12.5%) of the 96 patients. All of the patients presented with the Union for International Cancer Control (UICC) stage I or II disease, and none of the patients presented with UICC stage III or IV disease. All of the patients had undergone primary curative resection, and 17 (17.7%) patients had a positive surgical margin. If indicated, patients received either gemcitabine-based or 5-FU-based chemotherapy as first-line chemotherapy according to their clinical disease stage and performance status.

The immunohistochemical analysis of RASSF6 was performed on the 96 primary lesions of pancreatic ductal adenocarcinoma and 20 resected lesions of chronic pancreatitis. Figure 1 represents the immunohistochemical results of RASSF6 expression in patients with pancreatic ductal adenocarcinoma and chronic pancreatitis. RASSF6 was detected in carcinoma cells in the tumor tissues and was localized predominantly in the cytoplasm, without nuclear staining. In addition, islet cells showed an extremely strong staining of RASSF6 expression. Tumor cytoplasmic expression of RASSF6 was classified as negative (Figure 1A), strongly positive (Figure 1B) and weakly positive (Figure 1C). Of the 96 patients with pancreatic cancer, 51 (53.1%) showed negative RASSF6 expression, 29 (30.2%) had weak RASSF6 staining, and 16 (16.7%) displayed strongly positive staining. Among the 20 patients with chronic pancreatitis, strong RASSF6 expression and weak RASSF6 expression were observed in 80% (16/20) and 20% (4/20) of cases, respectively.

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Representative images of chronic pancreatitis and pancreatic ductal adenocarcinoma tissues immunostained for RASSF6. Representative micrographs of RASSF6 in pancreatic ductal adenocarcinoma (PDAC) cancerous tissues or chronic pancreatitis tissues by immunohistochemistry. A: Diffused cytoplasmic RASSF6 staining in chronic pancreatitis tissues; B: Representative negative staining of RASSF6 in PDAC; C: Extremely strong RASSF6 staining in pancreas islet (black arrow); D-F: Strongly positive RASSF6 staining in PDAC; G-I: Weakly positive RASSF6 staining in PDAC. Magnifications × 100 for A, B, D, E, G and H; × 200 for C, F and I.

Immunohistochemical analysis demonstrated a significantly higher proportion of positive RASSF6 staining in chronic pancreatitis compared with pancreatic cancer (Figure 1) (P = 0.015, Fisher’s exact test). In addition, qRT-PCR results revealed that the mRNA levels of RASSF6 were also down-regulated in PDAC compared with chronic pancreatitis (Figure 2A, P < 0.001). The relationship between the immunohistochemical stage of RASSF6 and its mRNA level was further evaluated (Figure 2B), and we found a clear difference in measured mRNA levels between negative, weak and strong RASSF6 staining groups. This finding indicates that the immunohistochemical scores are consistent with qRT-PCR results. Subsequently, RASSF6 mRNA expression and protein expression in 16 cases of PDAC samples with strong positive RASSF6 staining were evaluated using real-time PCR and Western blot analysis. The results confirmed that RASSF6 protein and mRNA levels were significantly down-regulated in cancer tissues (Figure 2C and D).

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Quantitative real-time polymerase chain reaction evaluation of RASSF6 expression in pancreatic ductal adenocarcinoma tissues. A: RASSF6 mRNA levels in pancreatic ductal adenocarcinoma (PDAC) tissues were significantly decreased compared with its levels in chronic pancreatitis tissues (P < 0.001); B: Groups were made according to the immunohistochemical results, and the consistency between immunohistochemical results and PCR results was evaluated through t test. Result showed that immunohistochemical scores were in keeping with RASSF6 mRNA levels (^bP < 0.001); C, D: RASSF6 mRNA and protein expression levels were evaluated by quantitative real-time polymerase chain reaction and Western blot in 16 cases with strong RASSF6 immunohistochemical staining. Even in these PDAC tissues with strongly positive RASSF6 staining, both the mRNA expression (C: P = 0.033) and protein levels (D: P = 0.049) were still less than para-tumor tissues.

Tumors with perineural invasion demonstrated less frequent RASSF6 expression, and a significant association between RASSF6 expression level and the presence of perineural invasion was observed (P = 0.026; Table 1). In addition, RASSF6 expression seemed to be associated with the UICC T stage. We found that higher-stage tumors demonstrated RASSF6 expression less frequently with a significant difference at the boundary (Table 1). No significant association was observed between age, sex, location of tumor, differentiation, lymph node status, portal vein invasion, or surgical margins.

Based on the immunohistochemical staining of RASSF6, 96 patients were then stratified into 3 groups (strongly positive vs weak vs negative). Kaplan-Meier curves were constructed to determine whether RASSF6 expression was correlated with survival in these groups. The median survival time of the strongly positive group, weak group and negative group was 33 mo, 15 mo and 11 mo, respectively. In this cohort, there was a significant correlation between RASSF6 expression levels and OS (P = 0.009; Figure 3A). Compared with patients with negative RASSF6A expression, patients with both strongly positive RASSF6 staining and weak RASSF6 staining showed improved OS time (strongly positive vs negative, P = 0.006; Figure 3B; weak vs negative, P = 0.047; Figure 3C). Furthermore, positive (weakly and strongly) RASSF6 expression was also strongly associated with an increased OS time compared with negative RASSF6 expression (P = 0.004) (Figure 3D).

Open in New Tab   Full Size Figure   Download Figure

Figure 3 Survival curves of patients with pancreatic ductal adenocarcinoma in regards to RASSF6 expression. A: There was a significant correlation between RASSF6 expression levels and overall survival (P = 0.009); B: Overall survival of patients with strongly positive RASSF6 staining was significantly higher than overall survival of those whose tumor was negative (P = 0.006); C: Weakly positive staining vs negative staining (P = 0.047); D: Overall survival of patients with positive RASSF6 staining (strongly positive plus weakly positive staining) remained better than survival of patients with negative staining of RASSF6 (P = 0.004).

The variables most likely to impact survival by univariate analysis (P≤ 0.1) were entered into the multivariate analysis model (Table 2). Cox regression analysis demonstrated that negative RASSF6 expression remained an independent prognostic factor of poor prognosis HR = 4.620, 95%CI: 2.027-10.527, P = 0.006).

Pancreatic ductal adenocarcinoma (PDAC) is a highly malignant digestive tumor with a very poor prognosis. Inappropriate activation of Ras proteins is a hallmark of PDAC, which promotes a variety of malignant phenotypes, including enhanced growth, loss of contact inhibition, reduced requirement for growth factors, enhanced motility and invasion and resistance to apoptosis.

Recently, RASSF6 was identified as a new negative effector of the RAS protein, which demonstrates a tumor suppressor function. In 239-T cells, a dramatic synergistic activation of cell death was observed when the cells were transfected with activated K-ras and RASSF6 together, and siRNA-mediated knockdown of RASSF6 expression in human lung tumor cells increases tumorigenicity. However, the role of RASSF6 in PDAC is still unknown. In this study, the authors demonstrated that RASSF6 expression is decreased in PDAC tissues and lymph node metastases, and the expression of RASSF6 has a significant impact on the survival of patients with PDAC. These results might indicate that RASSF6 could be a potential prognostic factor for PDAC patients.

This is the first study to report that RASSF6 is decreased in PDAC and that the expression of RASSF6 might decrease with the development of cancer.

This study showed that the RASSF6 expression level may be used as a potential predictor of prognosis in pancreatic cancer.

RASSF6 is the abbreviation of Ras association (RalGDS/AF-6) domain family member 6. This gene encodes a member of the Ras-association domain family (RASSF). Members of this family form the core of a highly conserved tumor suppressor network, the Salvador-Warts-Hippo pathway. The protein encoded by this gene is a Ras effector protein that induces apoptosis. A genomic region containing this gene has been linked to susceptibility to viral bronchiolitis. Alternative splicing results in multiple transcript variants and protein isoforms.

This is an interesting study with valuable information regarding the expression and clinical impact of RASSF6 in pancreatic ductal adenocarcinoma.